ITS N D

LIME Y O WHO WE ARE W H AT W E CA N D O WHERE WERE HEADED

B

The Great
ECO LO GY

Climate
Experiment
How far can we push the planet?
By Ken Caldeira
Illustrations by Tyler Jacobson

business, government or technology

forecasts usually look five or 10 years
out, 50 years at most. Among climate
scientists, there is some talk of centu-
rys end. In reality, carbon dioxide
dumped into the atmosphere today
will affect Earth hundreds of thou-
sands of years hence.
How will greenhouse gases change
the far future? No one can say for sure
exactly how Earth will respond, but cli-
mate scientistsusing mathematical
models built from knowledge of past
climate systems, as well as the complex
web of processes that impact climate
and the laws of physics and chemis-
trycan make predictions about what
Earth will look like.
Already we are witnessing the fu-
ture envisioned by many of these mod-
els take shape. As predicted, there has
been more warming over land than
over the oceans, more at the poles
than near the equator, more in winter
than in summer and more at night

78 Scientific American, September 2012 Photograph by Tktk Tktk

2012 Scientific American

 NEAR FUTURE: 
Industrial civilization
continues to pump out
more and more green
house gases with each
passing year, which will
result in hotter tempera
tures, an acidified ocean
and weirder weather
by centurys end.

2012 Scientific American

 ITS N D
LIME Y O WHO WE ARE
B
Ken Caldeira is a climate scientist
working for the Carnegie Institution
for Sciences Department of Global
Ecology at Stanford University. He
investigates issues related to climate,
carbon and energy systems. Caldeiras
primary tools are climate and carbon
cycle models, and he also does field
work related to ocean acidification.
80 Scientific American, September 2012
W H AT W E CA N D O WHERE WERE HEADED
than in the day. Extreme downpours have
become more common. In the Arctic, ice
and snow cover less area, and methane-
rich permafrost soils are beginning to
melt. Weather is getting weirder, with
storms fueled by the additional heat.
What are the ultimate limits of the
change that we are causing? The best his-
torical example comes from the 100-mil-
lion-year-old climate of the Cretaceous
period, when moist, hot air enveloped di-
nosaurs leathery skin, crocodilelike crea-
tures swam in the Arctic and teeming
plant life flourished in the CO2-rich air.
The greenhouse that is forming now will
have consequences that last for hundreds
of thousands of years or more. But first, it
will profoundly affect much of life on the
planetespecially us.
A DESERT IN ITALY
one of the greatest uncertainties in cli-
mate prediction is the amount of CO2 that
will ultimately be released into the atmo-
sphere. In this article, I will assume indus-
trial civilization will continue to do what
it has been doing for the past 200 years
namely, burn fossil fuels at an accelerat-
ing rate until we can no longer afford to
pull them out of the ground.
Just how much CO2 could we put into
the atmosphere? All told, there are about
one quadrillion metric tons (1021 grams)
of organic carbon locked up in Earths
sedimentary shell in one form or another.
So far we have burned only one twentieth
of 1 percent of this carbon, or roughly
2,000 billion metric tons of CO2.
With all the carbon locked in Earths
crust, we will never run out of fossil fuels.
We are now extracting oil from tar sands
and natural gas from water-fractured
shaleboth resources once thought to be
technologically and economically inac-
cessible. No one can confidently predict
just how far ingenuity can take us. Yet
eventually the cost of extraction and pro-
cessing will become so high that fossil fu-
els will become more expensive than al-
ternative resources. In the scenario envis-
aged here, we ultimately burn about 1
percent of the available organic carbon
over the next few centuries. That is in the
range of the amount of extraction most
likely to become technologically feasible
in the foreseeable future. We further as-
sume that in the future humanity will
learn to extract unconventional fossil fu-
els but will burn them at slower rates.
2012 Scientific American
Without any change in our habits,
Earth may warm by about five degrees
Celsius (nine degrees Fahrenheit) by 2100,
although the actual warming could be half
or even double this amount, depending
primarily on how clouds respond. This
change is about the difference between
the average climate of Boston, Mass., and
Huntsville, Ala.
In the northern midlatitudes between
30 degrees north and 60 degrees north
a band that includes the U.S., Europe,
China, and most of Canada and Russia
the annual average temperature drops
two thirds of a degree C with each degree
of increasing latitude. With five degrees C
of warming in a century, that translates
into an average poleward movement of
more than 800 kilometers in that period,
for an average poleward movement of
temperature bands exceeding 20 meters
each day. Squirrels may be able to keep
up with this rate, but oak trees and earth-
worms have difficulty moving that fast.
Then there will be the rains. Earth is a
planetary-scale heat engine. The hot sun
warms equatorial air, which then rises
and cools. The cooling condenses water
vapor in the air, which falls back to Earth
as rainhence, the belt of torrential rains
that occur near the equator.
Yet this water condensation also heats
the surrounding air, causing it to rise even
more rapidly. This hot, dry air reaches as
high as jets fly, then spreads laterally to-
ward the poles. At altitude, the hot air radi-
ates heat to space and thus becomes cool,
which causes it to sink back toward the
planets surface. The suns rays pass
through this dry, cloudless air, beating
down to heat the arid surface. Today such
dry air sinks occur at about 30 degrees
north and south latitude, thus creating the
great belts of desert that encircle the globe.
With greenhouse warming, the rising
air is hotter. Thus, it takes more time for
this air to cool off and sink back to Earth.
As a result, these desert bands move to-
ward the poles.
The climate of the Sahara Desert may
move northward. Already southern Eu-
rope has been experiencing more intense
droughts despite overall increases in pre-
cipitation globally, and it may lose the
Mediterranean climate that has long
been considered one of the most desir-
able in the world. Future generations may
say the same about the Scandinavian cli-
mate instead.
 Up there in the northern midlatitudes,
growing seasons are getting longer. Spring
springs sooner: plants flower, lake ice
melts and migratory birds return earlier
than in the historical past.
That will not be the only benefit to
croplands in Canada and Siberia. Plants
make food by using the energy in sunlight
to merge CO2 and water. For the most
part, plants absorb CO2 via little pores in
leaves known as stomata. When the sto-
mata are open wide, the plants can get
plenty of CO2, but a lot of water evaporates
through these gaping holes. Higher con-
centrations of atmospheric CO2 mean a
plant can get the CO2 it needs by opening
its stomata slightly or even building fewer
stomata in leaves. In a high-CO2 world,
plants can grow more using the same
amount of water. (This decrease in evapo-
ration from plants also leads to a further
decrease in precipitation, and because
evaporation causes cooling, the decrease
in evaporation causes further warming.)
Such gains will not be felt everywhere.
In the tropics, high temperatures already
compromise many crops; this heat stress
will likely get worse with global warming.
The outlook may be for increased crop
productivity overall, with increases in the
north exceeding the reductions near the
equator. Global warming may not de-
crease overall food supply, but it may give
more to the rich and less to the poor.
OCEANS OF CHANGE
the vast oceans resist change, but change
they will. At no time in Earths pastwith
the possible exception of mass-extinction
eventshas ocean chemistry changed as
much and as rapidly as scientists expect
it to over the coming decades. When CO2

FA S T - F O R WA R D

Climate: Past as Future

Assuming we continue to burn fossil fuels at will, releasing
greenhouse gases such as carbon dioxide unabated into the at
mosphere, the planet will be transformed. Already global temper
atures have risen by nearly one degree Celsiusmore than twice
that in the Arctic. Average temperatures could eventually rise by
10 degrees C, enough to melt the vast quantities of water stored
as ice in the glaciers of Greenland and Antarctica. Enough water
could be released to raise sea levels by 120 meters. Atmospheric
concentrations of carbon dioxide will reach levels last seen during
the Cretaceous period, when dinosaurs roamed Earth, North America
was cut in two by an enormous inland sea and crocodilelike creatures
inhabited the poles.

Global temperature
(annual average,
degrees Celsius) 25 14 15 20 25

High

Carbon dioxide
(atmospheric
concentration)

Sea level

Biomass

Biodiversity
Low
CRETACEOUS PREINDUSTRIAL TODAY NEAR FUTURE FAR FUTURE
(100 million years ago) (100 years ago) (100 years from now) (10,000 years from now)

Maps by XNR Productions, Graphic by Jen Christiansen September 2012, ScientificAmerican.com81

2012 Scientific American
 ITS N D
LIME Y O WHO WE ARE
B
FAR FUTURE: If green
house gas emissions from
burning fossil fuels continue
unabated, sea levels may rise
by 120 meters and polar
regions will become much
warmer. Any human civiliza
tion still extant will need to
adapt to these conditions.
82 Scientific American, September 2012
W H AT W E CA N D O WHERE WERE HEADED
enters the oceans, it reacts with seawater
to become carbonic acid. In high enough
concentrations, this carbonic acid can
cause the shells and skeletons of many
marine organisms to dissolveparticular-
ly those made of a soluble form of calcium
carbonate known as aragonite.
Scientists estimate that more than a
quarter of all marine species spend part of
their lives in coral reefs. Coral skeletons are
made of aragonite. Even if chemical condi-
tions do not deteriorate to the point where
shells dissolve, acidification can make it
more difficult for these organisms to build
them. In just a few decades there will be
no place left in the ocean with the kind of
chemistry that has supported coral-reef
growth in the geologic past. It is not known
how many of these coral-dependent spe-
cies will disappear along with the reefs.
2012 Scientific American
Such chemical changes will most di-
rectly affect reef life, but the rest of us
would be wise to consider the physical
changes afoot. At the most basic level,
water acts like mercury in a thermometer:
add heat and watch it rise. The sea is also
being fed by water now held in ice caps.
In high-CO2 times in the ancient past,
Earth warmed enough for crocodilelike
animals to live north of the Arctic Circle.
Roughly 100 million years ago annual av-
erage polar temperatures reached 14 de-
grees C, with summertime temperatures
exceeding 25 degrees C. Over thousands
of years temperatures of this magnitude
would be sufficient to melt the great ice
sheets of Greenland and Antarctica. With
the ice sheets melted completely, sea level
will be about 120 meters higher, flooding
vast areas. That waters weight on low-
lying continental regions will push those
areas down farther into the mantle, caus-
ing the waters to lap even higher.
The poles are expected to warm about
2.5 times faster than Earth as a whole. Al-
ready the Arctic has warmed faster than
anywhere else, by about two degrees C
compared with 0.8 degree C globally. At
the end of the last ice age, when the cli-
mate warmed by about five degrees C over
thousands of years, the ice sheets melted
at a rate that caused sea level to rise about
one meter per century. We hope and ex-
pect that ice sheets will not melt more rap-
idly this time, but we cannot be certain.
CHASING VENUS
over the past several million years Earths
climate has oscillated to cause the waxing
and waning of great ice sheets. Our green-
house gas emissions are hitting this com-
plex system with a hammer. I have pre-
sented a scenario in which our climate
evolves fairly smoothly, but jumps and
starts that could shock biological, social
and political systems beyond the limits of
their resilience are also possible.
Consider that Arctic warming could
cause hundreds of billions of metric tons
of methane to rapidly bubble to the atmo-
sphere from Arctic seabeds and soils. Mol-
ecule for molecule in the atmosphere,
methane is about 37 times better at trap-
ping heat than CO2. Were this methane
released suddenly, as may have occurred
in a warming event 55 million years ago
known as the Paleocene-Eocene Thermal
Maximum, we could experience truly cat-
astrophic warming. This risk is remote,
however, according to most scientists.
Some have also suggested that feed-
back effects such as melting permafrost
could cause a runaway greenhouse scenar-
io where the oceans become so hot they
evaporate. Because water vapor is itself a
greenhouse gas, such a stronger water cy-
cle could cause Earth to get so hot that at-
mospheric water vapor would persist and
never rain out. In this case, atmospheric
CO2 from volcanoes and other sources
would continue to accumulate. Cosmic
rays would break apart the water vapor at
high altitudes; the resulting hydrogen
would eventually escape to space. Earths
climate would then settle into a state rem-
iniscent of its planetary neighbor Venus.
Fortunately, ocean vaporization is not
even a remote risk from todays green-
house gas emissions. Simply put, there is a
limit to how much CO2 can heat the plan-
et. Once CO2 and water vapor concentra-
tions rise high enough, the molecules in-
creasingly scatter the incoming sunlight,
preventing it from getting any hotter.
If we continue to burn fossil fuels,
however, greenhouse gas concentrations
in the atmosphere will reach levels last
seen in the Cretaceous. Back then, inland
seas flooded vast areas of the continents
on a hot, moist Earth. Giant reptiles swam
in the oceans. On land, dinosaurs grazed
on luxuriant plant growth. If we burn
just 1 percent of the organic carbon in
Earths crust over the next few centuries,
humans will breathe the same CO2 con-
centrations as the dinosaurs inhaled and
experience similar temperatures.
Compared with the gradual warming
of hothouse climates in the past, industrial
climate change is occurring in fast-for-
ward. In geologic history, transitions from
low- to high-CO2 atmospheres typically
happened at rates of less than 0.00001 de-
gree a year. We are re-creating the world of
the dinosaurs 5,000 times faster.
What will thrive in this hothouse? Some
organisms, such as rats and cockroaches,
are invasive generalists, which can take ad-
vantage of disrupted environments. Other
organisms, such as corals and many tropi-
cal forest species, have evolved to thrive in
a narrow range of conditions. Invasive spe-
cies will likely transform such ecosystems
as a result of global warming. Climate
change may usher in a world of weeds.
Human civilization is also at risk. Con-
sider the Mayans. Even before Europeans
arrived, the Mayan civilization had begun
to collapse thanks to relatively minor cli-
mate changes. The Mayans had not devel-
oped enough resilience to weather small
reductions in rainfall. The Mayans are
not alone as examples of civilizations that
failed to adapt to climate changes.
Crises provoked by climate change are
likely to be regional. If the rich get richer
and the poor get poorer, could this set in
motion mass migrations that challenge
political and economic stability? Some of
the same countries that are most likely
to suffer from the changes wrought by
global warming also boast nuclear weap-
ons. Could climate change exacerbate ex-
isting tensions and provoke nuclear or
other apocalyptic conflict? The social re-
sponse to climate change could produce
bigger problems for humanity than the
climate change itself.
2012 Scientific American
STARTING OVER
the woody plants that flourished during
the Cretaceous died, and some became coal
over geologic time. The oceans plankton
ended up buried in sediments, and some
became oil and gas. The climate cooled as
sea life locked CO2 in shells and skeletons.
The oceans will absorb most of our CO2
over millennia. The resulting acidification
will dissolve carbonate minerals, and the
chemical effects of dissolution will allow
yet more CO2 to be absorbed. Neverthe-
less, atmospheric CO2 concentrations will
remain well above preindustrial levels of
280 parts per million for many tens of
thousands of years. As a result, the ebb
and flow of ice ages brought on by subtle
variations in Earths orbit will cease, and
humanitys greenhouse gas emissions will
keep the planet locked in a hothouse.
Over time increased temperatures and
precipitation will accelerate the rate at
which bedrock and soils dissolve. Streams
and rivers will bring these dissolved rocks
and minerals, containing elements such
as calcium and magnesium, to the oceans.
Perhaps hundreds of thousands of years
from now some marine organism will
take the calcium and CO2 and form a car-
bonate shell. That seashell and millions of
others may eventually become limestone.
Just as the White Cliffs of Dover in Eng-
land are a remnant of the Cretaceous at-
mosphere, the majority of carbon in the
fossil fuels burned today will become a lay-
er in the rocksa record, written in stone,
of a world changed by a single species.
MORE TO EXPLORE
Oceanography: Anthropogenic
Carbon and Ocean pH. Ken Caldeira
and Michael E. Wickett in Nature, Vol.
425, page 365; September 25, 2003.
Climate Change 2007: The Physi-
cal Science Basis. International
Panel on Climate Change. Cam-
bridge University Press, 2007. 
www.ipcc.ch/publications_and_
data/ar4/wg1/en/contents.html
The Long Thaw. David Archer.
Princeton University Press, 2010.
SCIENTIFIC AMERICAN
ONLINE
For more on what to expect from
climate change, watch the video at
ScientificAmerican.com/sep2012/
future-climate
September 2012, ScientificAmerican.com83